Randomizing Policy and the Risk of Multiple Hypothesis Testing

Randomized policy research should be pre-registered to reduce the chance of false positive results.

The Regulatory Review